remove xrt from oneflow

[hjchen2]

We have refactored XRT to a standalone project named OneFlow-XRT（https://github.com/Oneflow-Inc/oneflow-xrt ）that is modular and easier-to-use. Welcome to experience and make suggestions.

[github-actions]

Code got formatted by CI. Please request CI again if you still want to have this PR merged. If the PR is from a forked repo, please download the patch files from the GitHub Actions web page and apply them locally.

[github-actions]

CI failed when running job: cuda-benchmark. PR label automerge has been removed

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8273/

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8273/

[github-actions]

Speed stats:

GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 130.4ms (= 13039.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 144.5ms (= 14454.8ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 144.5ms / 130.4ms)

OneFlow resnet50 time: 77.0ms (= 7704.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.3ms (= 8526.8ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 85.3ms / 77.0ms)

OneFlow resnet50 time: 53.9ms (= 10778.1ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.4ms (= 11870.7ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.10 (= 59.4ms / 53.9ms)

OneFlow resnet50 time: 43.2ms (= 8648.0ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 44.4ms (= 8874.3ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.03 (= 44.4ms / 43.2ms)

OneFlow resnet50 time: 37.5ms (= 7492.3ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 36.9ms (= 7374.1ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.98 (= 36.9ms / 37.5ms)

OneFlow swin dataloader time: 0.240s (= 48.070s / 200, num_workers=1)
PyTorch swin dataloader time: 0.148s (= 29.536s / 200, num_workers=1)
Relative speed: 0.614 (= 0.148s / 0.240s)

OneFlow swin dataloader time: 0.100s (= 20.045s / 200, num_workers=4)
PyTorch swin dataloader time: 0.045s (= 9.027s / 200, num_workers=4)
Relative speed: 0.450 (= 0.045s / 0.100s)

OneFlow swin dataloader time: 0.036s (= 7.260s / 200, num_workers=8)
PyTorch swin dataloader time: 0.023s (= 4.626s / 200, num_workers=8)
Relative speed: 0.637 (= 0.023s / 0.036s)

❌ OneFlow resnet50 time: 148.0ms (= 14795.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 172.8ms (= 17284.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 172.8ms / 148.0ms)

OneFlow resnet50 time: 97.9ms (= 9792.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 112.3ms (= 11231.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 112.3ms / 97.9ms)

OneFlow resnet50 time: 74.4ms (= 14871.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 98.9ms (= 19787.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.33 (= 98.9ms / 74.4ms)

OneFlow resnet50 time: 59.7ms (= 11947.1ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.7ms (= 14940.6ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.25 (= 74.7ms / 59.7ms)

OneFlow resnet50 time: 54.3ms (= 10868.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 68.9ms (= 13789.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 68.9ms / 54.3ms)